Game bidding strategy for electric thermal gas coupled virtual power plants participating in main and auxiliary joint markets

Authors

• Chunyan ZHANG, State Grid Shanghai Municipal Electric Power Company, Shanghai 200023, China
• Zhenlan DOU, State Grid Shanghai Municipal Electric Power Company, Shanghai 200023, China
• NG Jihang ZHA, School of Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
• Lingling WANG, School of Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, ChinaFollow
• Chuanwen JIANG, School of Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China

Keywords

multi-energy coupled virtual power plant; constraint-aware; multi-agent reinforcement learning; main and auxiliary joint markets; game bidding strategy

Abstract

As integrated energy systems develop, multi-energy coupled virtual power plants have become an effective way to manage integrated energy systems. However, due to their numerous internal energy types, complex coupling relationships, and inability to obtain competitor information when participating in the market, it is difficult to solve their bidding strategies by adopting traditional mathematical methods. A game bidding strategy for electric thermal gas coupled virtual power plants participating in main and auxiliary joint markets based on constraint-aware multi-agent reinforcement learning is proposed. Firstly, a mathematical model of a multi-energy coupled virtual power plant is built and the energy flow relationship of its internal electric thermal gas multi-energy coupling is described. Secondly, a game model of multi-energy coupled virtual power plants is built, in which price setters participate in the incomplete information main and auxiliary joint market, and virtual power plants are allowed to trade electricity and auxiliary services. Finally, constraint-aware reinforcement learning is improved, and the constraint-aware multi-agent reinforcement learning algorithm is proposed for solving the model. By setting a watchdog module in the algorithm, the virtual power plant strategy is effectively ensured to comply with power grid safety constraints, and compared with the theoretical optimal solution to verify the effectiveness of the proposed method.

DOI

10.19781/j.issn.1673-9140.2026.02.014

First Page

156

Last Page

166

Recommended Citation

ZHANG, Chunyan; DOU, Zhenlan; ZHA, NG Jihang; WANG, Lingling; and JIANG, Chuanwen (2026) "Game bidding strategy for electric thermal gas coupled virtual power plants participating in main and auxiliary joint markets," Journal of Electric Power Science and Technology: Vol. 41: Iss. 2, Article 14.
DOI: 10.19781/j.issn.1673-9140.2026.02.014
Available at: https://jepst.researchcommons.org/journal/vol41/iss2/14